DESCRIPTION: (Applicant's Abstract) Non-small cell lung cancer (NSCLC) is poorly responsive to current chemotherapeutic regimens with an overall regression rate of only 30-50%. The mechanisms of resistance are likely to involve multiple gene products. In preliminary studies the H1435 non-small cell lung cancer (NSCLC) cell line is 50-fold more resistant to carboplatin than H460. Evaluation of 20 genes putatively associated with chemoresistance revealed that glutathione transferase (GST) P1, Bax alpha, GADD45, ERCC3, and glutathione peroxidase, mGST genes are expressed at 100, 20, 10, 6, 5, and 4-fold higher levels respectively in H1435. The primary long term objective of the proposed investigation is to improve mechanistic understanding of NSCLC chemoresistance and to develop a method for predicting which NSCLC tumors will respond to carboplatin. Achieving these goals will have specific value by 1) allowing development of improved chemotherapeutic agents based on a knowledge of mechanisms, 2) sparing patients with carboplatin resistant tumors unnecessary suffering and expense by avoiding an ineffective treatment, 3) allowing patients with carboplatin resistant tumors to pursue alternative treatment options without delay, and 4) providing a general model for identifying tumors resistant to other chemotherapeutic regimens. To achieve these goals, many genes must be evaluated. Carcinoma cell lines will be the most appropriate model for a comprehensive investigation. The specific aims are the following. Aim 1) Determine expression of eighty putative carboplatin chemoresistance genes in ten NSCLC lines, determine the IC30 concentration of carboplatin for each line, then determine which constitutive expression patterns correlate best statistically (by linear regression or by Duncan's Multiple Range test) with resistance. Any correlation identified will be validated in ten additional NSCLC lines. Aim 2) Combine the individual constitutive gene expression values into indices and assess them for association with the initial ten chemoresistant cell lines. Any correlation will be validated in the second ten lines. The competitive template RT-PCR gene expression quantification method proposed for this project is suggested by the applicant to be the best available because 1) it allows high throughput measurement of many samples. 2) the data produced are in standard units so that individual gene expression values may be combined mathematically in gene expression indices, and 3) because the same standard mixture of competitive templates will be used for all of the proposed experiments, data from all experiments may be compared for statistical analysis.